A multicellular model of mammalian circadian rhythm generation and synchronization will be developed through an integrated program ofneurophysiological experiments, theoretical modeling and multi-scale systems analysis and computation. Our hypothesis is that the coupling of individual pacemaker neurons is mediated by the neurotransmitter vasoactive intestinal peptide (VIP) prevalent in the suprachiasmatic nucleus (SCN). Experiments will identify the dose- and phase-dependence of SCN neurons on VIP for circadian synchrony. These experiments will guide the development of a pacemaker cell model in which a detailed description of the gene regulatory network responsible for rhythmic electrical activity is combined with a simplified description of the VIP signaling pathways implicated in circadian coupling. The pacemaker model will be used as a building block in the construction of neural population models that account for the known anatomy and physiology of the SCN including the core and shell divisions and the distribution of VIP producing cells in the two divisions. The resulting models will cover a wide range of time and length scales ranging from the gene regulation level to the neuron signaling level to the tissue level. Deterministic simulation codes will be developed to allow the efficient simulation of large ensembles of coupled SCN neurons. Stochastic effects at the gene and signaling levels will be studied by developing combined deterministic/stochastic simulation codes. Parallel theoretical work on the population model and a simplified surrogate model will yield insights into stochastic effects in large neuron populations. The combined experimental, theoretical and computational work will allow systematic perturbation analysis of the roles of individual neurons and their interconnections on the precision and robustness of circadian rhythm generation.